REPRESENTING IGNORANCE*
By Russell Hardin
I. Introduction
If we wish to assess the morality of elected officials, we must understand their function as our representatives and then infer how they can
fulfill this function. I propose to treat the class of elected officials as a
profession, so that their morality is a role morality and it is functionally
determined. If we conceive the role morality of legislators to be analogous
to the ethics of other professions, then this morality must be functionally
defined by the purpose that legislators are to fulfill once in office. Hence,
the role morality of legislators will largely be determined by our theory of
representation. We will need not a normative account of their role, but an
empirical explanatory account. In David Hume’s terms, the morality of
role holders is one of “artificial” duties, that is to say, duties defined by
their functional fit with the institutional purposes of a profession.1 Our
most difficult problem, therefore, is to understand the role of our elected
representatives.
This problem is severely complicated by the nature of democratic choice
and participation in a modern, complex society. A central problem of
democratic theory for such a society is the general political ignorance of
the citizens. In Capitalism, Socialism, and Democracy (1942), Joseph Schumpeter argues that citizens have no chance of affecting electoral outcomes
and, therefore, no reason to learn enough about politics even to know
which candidates or policies would serve their interests. He writes, “[W]ithout the initiative that comes from immediate responsibility, ignorance
will persist in the face of masses of information however complete and
correct.” 2 If the problem of knowing enough to judge elected government
officials is already hard, the lack of incentive to correct this problem is
* I am indebted to my fellow contributors to this volume, to participants in the Monday
Night Theorists at New York University, and to Ellen Frankel Paul for comments on an
earlier draft of this essay.
1
Russell Hardin, “The Artificial Duties of Contemporary Professionals,” Social Service
Review 64 (1991): 528–43. Hume distinguishes “natural” from “artificial” virtues in David
Hume, A Treatise of Human Nature (1739–40), ed. L. A. Selby-Bigge and P. H. Nidditch, 2d ed.
(Oxford: Oxford University Press, 1978), bk. 3, pt. 2, sec. 1. Acting from a natural duty
produces a good directly, and more or less immediately. Acting from an artificial duty, such
as the duty of justice, produces good only through the mediation of a social institution or
norm.
2
Joseph A. Schumpeter, Capitalism, Socialism, and Democracy (1942), 3d ed. (New York:
Harper, 1950), 262.
76
© 2004 Social Philosophy & Policy Foundation. Printed in the USA.
REPRESENTING IGNORANCE
77
devastating. Indeed, the costs of knowing enough about government to
be able to vote intelligently in one’s own interest surely swamp the modest costs, for most people in the United States and other advanced democratic nations, of actually casting a vote, at least on commonplace issues
of public policy outside moments of great crisis. Therefore, an economic
theory of knowledge or “street-level epistemology” weighs against knowing enough to vote well because the incentives cut heavily against investing in the relevant knowledge.3 The typical voter will not be able to put
the relevant knowledge to beneficial use.4
One response to the problem of citizens’ incompetence to judge how
they should be governed is government by Burkean representatives.
Throughout his writings, Edmund Burke (1729–97) supposes that only
members of an elite are competent to govern, and the mass of the citizenry ought to turn government over to them to do what they think best.5
I will suppose that we resort to representative government for a very
different, structural reason. It is impossible for the entire polity of a large
state to make policy directly or to implement it.6 We therefore have specialized bodies to do these things. It is this structural fact that wrecks our
incentive to know much about the politics that governs much of our lives.
Still, we want representative government genuinely to represent us, to
adopt the policies that we would adopt if we had relevant knowledge and
power. Any actual representative government may have many problems
that get in the way of its serving us in such a manner,7 but I will ignore
those problems and focus on the ethics of the representative in trying to
3
Russell Hardin, “The Street-Level Epistemology of Democratic Participation,” Journal of
Political Philosophy 10, no. 2 (2002): 212–29; reprinted in James Fishkin and Peter Laslett, eds.,
Philosophy, Politics, and Society, vol. 7 (London: Blackwell, 2003), 163–81. The argument here
is essentially from the logic of incentives and it casts no blame on anyone. Others hold the
political system, the poor educational system, or politicians responsible for the seeming
incompetence of voters. See, for example, Matthew A. Crenson and Benjamin Ginsberg,
Downsizing Democracy: How America Sidelined Its Citizens and Privatized Its Public (Baltimore,
MD: Johns Hopkins University Press, 2002).
4
Martin P. Wattenberg argues that, additionally, there are obstacles even to casting one’s
vote correctly in the United States, where the act of voting can be almost as difficult as taking
a college entrance examination. See Martin P. Wattenberg, Where Have All the Voters Gone?
(Cambridge, MA: Harvard University Press, 2002), chap. 6.
5
See, for example, Edmund Burke, “Speech to the Electors of Bristol” (1774), in Hanna
Fenichel Pitkin, ed., Representation (New York: Atherton Press, 1969), esp. 174–75; and Edmund Burke, “Appeal from the New to the Old Whigs” (1791), in The Works of Edmund Burke
(London: George Bell & Sons, 1890–1906), 3:85–87. For later views, see Michael Oakeshott,
Rationalism in Politics and Other Essays (New York: HarperCollins, 1962), esp. Oakeshott’s
essay, “On Being a Conservative.”
6
Even in a not so modern polity —England and Wales in 1754 —the electorate was about
two hundred and eighty thousand out of a population of roughly eight million. See Bernard
Manin, The Principles of Representative Government (Cambridge: Cambridge University Press,
1997), 82. In the United States at the time of ratification of its constitution in 1788, the
electorate was about one million out of a population of a little more than three million. Thus,
even these relatively small populations, by modern standards, required representative government or some other device for making decisions without involving all citizens at once.
7
Ibid.
78
RUSSELL HARDIN
represent others in the face of both gross ignorance on the part of voters
and harsh limits on information for the representative.
I will not assume the position of those who view representative democracy as a forum for deliberation on “the truth.” In The Principles of
Representative Government (1997), Bernard Manin states that political scientist Carl Schmitt (1888–1985) holds this view, namely, “that truth must
‘make the law,’ ” and that “debate is the most appropriate means of
determining truth, and therefore the central political authority must be a
place of debate, that is, a parliament.” 8 I will not deal with this vision
here, because I think it largely irrelevant. Parliamentary debate does not
often approach “truth,” and, in any case, politics is far more about interests than it is about truth.9 If there is a truth to be discovered and demonstrated, then after sufficient debate there should be consensus and each
of us should grasp the truth. In that case, the Schmitt thesis would make
sense. When there is a conflict of interests, however, there is no truth of
the kind Schmitt envisions. Deliberation on interests is as likely to lead to
dissensus as to consensus.10 I should also note that Schmitt’s justification
of a parliamentary body does not require that the body actually be representative of the various groups in a society, but only that it be the
political decision-making body for the society.
In general, the tasks of representatives will be easiest when they merely
represent interests of fairly basic kinds on issues, for example, of economic policy or welfare provision. Their tasks might also seem to be
straightforward when they deal with hotly contested issues over which
public views are relatively forcefully asserted, as with the contemporary
debate over abortion in the United States. Indeed, for any controversial
matter such as abortion, representatives might often think of themselves
as delegates because they would not expect to get reelected if they reneged on their electioneering commitment to take one side or the other on
the issue (although there might be intermediate positions on which they
could compromise). The most difficult issues for representatives will generally be those for which clarity is lacking, at least in the sense that
citizens do not know where they stand or, rather, where their interests lie.
As John Stuart Mill says, individuals distort many issues with their idiosyncratic beliefs, and they tend to discount the future, so that individuals’ claims of what their interests are can be distorted.11
8
Ibid., 185. See also Carl Schmitt, The Crisis of Parliamentary Democracy (1923), trans. Ellen
Kennedy (Cambridge, MA: MIT Press, 1988), 35, 43.
9
Geoffrey Brennan and James M. Buchanan, The Reason of Rules: Constitutional Political
Economy (Cambridge: Cambridge University Press, 1985); and Russell Hardin, “Deliberation: Method, Not Theory,” in Stephen Macedo, ed., Deliberative Politics: Essays on Democracy
and Disagreement (New York: Oxford University Press, 1999), 103–19.
10
See Hardin, “Deliberation: Method, Not Theory”; and Cass Sunstein, “Deliberative
Trouble? Why Groups Go to Extremes,” Yale Law Journal 110 (2000): 71–119.
11
John Stuart Mill, Considerations on Representative Government (1861), in John Stuart Mill,
Essays on Politics and Society, vol. 19 of Collected Works of John Stuart Mill, ed. J. M. Robson
(Toronto: University of Toronto Press, 1977), 444–45.
REPRESENTING IGNORANCE
79
II. Voter Ignorance
If, in general, we make the effort to know something in large part
because we think it will serve our interest to know it, then we cannot expect
people to know very much about what their representatives do. In the
argument of the “economic theory of democracy,” a citizen typically does
not have very much interest in voting. One vote has a miniscule chance
of making a difference, so miniscule that, even when it is multiplied by
the value of making a difference and getting one’s preferred candidate or
policy, the expected value of the vote is vanishingly slight. Therefore, if
there is any real cost involved in casting a vote, this cost swamps the
expected benefit to the voter of voting. Hence, there is little point in
knowing enough actually to vote well.
Most of the research on and debate over voting since Schumpeter’s
Capitalism, Socialism, and Democracy and Anthony Downs’s An Economic
Theory of Democracy (1957)12 has focused primarily on the incentive to
vote rather than the incentive to know enough to vote intelligently.13 The
latter is at least logically derivative from the former, because it is the lack
of incentive to vote that makes knowledge of how to vote well virtually
useless, so that mastering such knowledge violates the pragmatic understanding of knowledge. Since my vote has miniscule causal effect on
democratically determined outcomes, there is no compelling reason for
me to determine how to vote intelligently. Or, to put this the other way
around, the fact that I would benefit from policy X does not give me
reason or incentive to know about or to understand the implications of
policy X unless I can somehow affect whether policy X is to be adopted.
In what follows, I will simply take for granted that typical citizens do
not master the facts they need to know if they are to vote their interests
intelligently. There is extensive evidence on this claim, although there is,
of course, also great difference of opinion on its significance for electoral
choices. For example, political scientist Samuel Popkin canvasses problems of voter ignorance in American presidential elections and then refers
to “low-information rationality,” which is rationality despite abysmal factual ignorance.14 He also argues for a ‘Gresham’s law’ of political information: bad facts drive out good facts. According to this law, “a small
amount of personal information [on a candidate] can dominate a large
amount of historical information about a past record.” 15 The personal
information might be some minor fact that comes up during a campaign.
However, the effect of Gresham’s law might often run against Popkin’s
claim for low-information rationality by driving out the little bit of policy12
Anthony Downs, An Economic Theory of Democracy (New York: Harper, 1957).
Hardin, “The Street-Level Epistemology of Democratic Participation.” (Passages in the
next two pages are taken from pages 217–19 of this essay.)
14
Samuel L. Popkin, The Reasoning Voter: Communication and Persuasion in Presidential
Campaigns, 2d ed. (Chicago, IL: University of Chicago Press, 1994).
15
Ibid., 73.
13
80
RUSSELL HARDIN
relevant information the voter has. If the personal information seems to
be politically relevant, however, its dominance over other information
might not matter so much. For example, Italy’s prime minister, Silvio
Berlusconi, has been caught up in an ongoing scandal over accusations
that he bribed his way to wealth and, thence, to political power. Sociologist Renato Mannheimer says, “Italians don’t understand the contents of
the processes against him, so they make their judgments according to
their political leanings.” 16 But some voters might judge him for his use of
office to serve his own interests.17
Manin argues that the form representative democracy now takes is
“audience democracy.” 18 The trouble with the large amount of historical
information that is, at least in principle, available is that voters do not
typically know much about it because it would be silly for them to invest
the time needed to acquire such information. Hence, in this age of media
and celebrity, we vote for personalities rather than for policies and thereby
give an outsized role to personal information.
As evidence of how seldom voters even seek better information before
voting, consider the difficulty that candidates have in getting their message across to voters. Congressional scholar Richard Fenno elegantly portrays the burden that candidates for the U.S. House of Representatives
face in merely finding people to talk to.19 Even professional political
scientists, who have a strong interest in knowing more about politics than
their mere interest in the outcome of elections would suggest, find it hard
to keep up with much of what happens. Weekly tallies of votes in the U.S.
House of Representatives and Senate, for example, are reported in some
newspapers, but with such brevity that only specialists on a particular
issue would find them meaningful.
Results of referenda on even relatively simple issues suggest astonishing misunderstanding by voters. California voters displayed cavalier irresponsibility in a 1994 referendum (Proposition 184) on a so-called threestrikes sentencing law that mandates harsh minimum prison terms for
repeat offenders.20 Consider two early cases to which the law was applied. The first involved a thief with a prior record, who was sentenced to
a term of twenty-five years to life, with no possibility of parole before
serving at least twenty years, for his “felony petty theft” of one slice of
pizza.21 In the second case, Russell Benson was sentenced to a similar
16
Quoted in Frank Bruni, “Italy, a Land of Tolerance, Even to Prime Ministers,” New York
Times, May 28, 2003, late edition–final, sec. A, p. 4, col. 3.
17
He pushes legislation that would directly benefit him as an oligopolistic media owner.
See “Italian Leader Faces Dissent over Control of the Media,” New York Times, April 6, 2003,
late edition–final, sec. A, p. 4, col. 1.
18
Manin, The Principles of Representative Government, 218–32.
19
Richard F. Fenno, Jr., Home Style: House Members in Their Districts (Boston, MA: Little,
Brown, 1978).
20
Susan Estrich, Getting Away with Murder: How Politics Is Destroying the Criminal Justice
System (Cambridge, MA: Harvard University Press, 1998).
21
“25 Years for a Slice of Pizza,” New York Times, March 5, 1995, sec. 1, p. 21.
REPRESENTING IGNORANCE
81
term for shoplifting a twenty-dollar carton of cigarettes.22 The threestrikes referendum was provoked by some truly gruesome crimes, yet it
is so badly framed that it brutalizes petty felons.
California voters also apparently displayed complete misunderstanding of a 1998 referendum (Proposition 3) to undo a prior referendum
(Proposition 198 of 1996) on open primaries. The prior referendum, passed
by the voters in presumable ignorance of its consequences, would stupidly have disallowed California representation at the national Republican and Democratic Party nominating conventions in the year 2000. After
the electorate failed to pass Proposition 3, administrative devices were
used to enable the state to distinguish Democratic and Republican voters
in its presidential primary elections, thereby securing representation at
the two party conventions.23 In this failure to enact Proposition 3, democracy was a charade and, when it failed due to ignorance and widespread
misunderstanding, a knowledgeable bureaucrat, California Secretary of
State Bill Jones, intervened against the democratic result. (In a subsequent
suit by the major parties against the law, the U.S. Supreme Court overturned Proposition 198 as a violation of the constitutional right of the
political parties to assemble.) 24
III. Madisonian Representatives
Edmund Burke supposes that constituents are apt not to understand
what would serve the public good and, therefore, he suggests that representatives should act in the true interest of the public rather than as
their constituents might want them to act.25 He further supposes that
constituents should recognize the superiority of their representatives to
carry out the task of serving the public interest. His view is not contrary
to Schmitt’s, and it would allow the selection of members of a parliament
by criteria other than representativeness of the polity. On this view, all
representatives represent everyone, or the public as a whole. Virtually no
one supposes this today, however, and from U.S. constitutional debates
forward, we generally assume that representatives primarily represent
merely the interests of their constituencies and of those who supported
their campaigns.
Let us drop two of Burke’s views: that representatives ought to be
concerned with the overall public good and that representatives are or
should be superior individuals, imbued with character and values that
only hereditary elites can acquire. A national party or a national executive
might claim to fulfill the first, although parties are invariably partisan and
national executives are commonly partisan. On the second, contrary to
22
Elisabeth R. Gerber et al., Stealing the Initiative: How State Government Responds to Direct
Democracy (Upper Saddle River, NJ: Prentice Hall, 2001), 64.
23
Ibid., 71–74.
24
California Democratic Party v. Jones, 530 U.S. 567 (2000).
25
See also Mill, Considerations on Representative Government, 511–12.
82
RUSSELL HARDIN
what Burke wants, what might set legislators apart is that they become
competent at politics, legislation, and governance through their specialized roles. Our representatives even tend, in Manin’s characterization, to
become aristocratic in that they must have relatively high levels of competence and achievement to attain and hold their offices.26 They clearly
do not represent their constituents in the sense of being like them. There
are, for example, almost no working-class representatives in modern democratic governments, and lawyers are radically overrepresented in U.S.
legislative bodies. Representation of groups must often be through socalled active representation by people who themselves do not directly
share the interests of the groups they represent. For example, Senator Ted
Kennedy (D–MA) often represents the interests of union members and
the poor, although he has no experience of either status in his own life.
An obvious but painful implication of the Schumpeterian world, in
which the public exhibits a general political ignorance, is that representatives can take advantage of citizens. This is true not merely in the
manner of Italy’s Berlusconi, who has used his official power to enact
laws that specifically benefit him by helping him avoid trial for bribery,
an offense that he has implicitly admitted.27 It is true more fundamentally
in the sense that, even without such overt actions as Berlusconi’s, government personnel can be parasitic on the larger society, making themselves wealthier than they otherwise could have been in any other
profession, giving themselves prerogatives far beyond their ordinary emoluments, and securing long tenure for themselves and often even their
relatives. In a sense well beyond Manin’s, they become an aristocratic
class apart from the society that they both govern and represent.
Even the slightest Madisonian or Humean view of human nature as
self-interested yields this implication. Political sociologist Robert Michels
(1876–1936) claims that democratic government within political parties —
especially European socialist parties —produced an aristocracy with great
power over rank-and-file members.28 This claim is true more generally of
democratic governments, although they may typically be subjected to
greater scrutiny that might impede some of the worst excesses of oligarchic power. In Michels’s famous slogan, “Who says organization says
oligarchy.” Perversely, who says representative democracy evidently also
says oligarchy.
26
Manin, The Principles of Representative Government, chap. 4.
Members of Berlusconi’s party in the Italian Parliament introduced legislation to exempt the top five government officials from facing trial while they hold office ( Jason Horowitz, “World Briefing/Europe: Berlusconi Immunity Plan,” New York Times, May 30, 2003,
late edition–final, sec. A, p. 8, col. 5). Berlusconi previously pushed through legislation to
reform the courts in ways that might have permitted him to avoid prosecution for bribery,
complaining that “to search for, and single out, individual culprits [is] disingenuous and
inherently unjust” when there is so much suspicious activity to go around (Bruni, “Italy, a
Land of Tolerance”).
28
Robert Michels, Political Parties: A Sociological Study of the Oligarchal Tendencies of Modern
Democracy (1911), trans. Eden and Cedar Paul (Glencoe, IL: Free Press, 1949).
27
REPRESENTING IGNORANCE
83
According to the Manin and Michels theses of an aristocracy of leadership, in some sense it is not the individual elected officials but the class
of them that is problematic. As John C. Calhoun (1782–1850) writes: “The
advantages of possessing the control of the powers of the government,
and thereby of its honors and emoluments, are, of themselves, exclusive
of all other considerations, ample to divide . . . a community into two
great hostile parties.” 29 As a class, the political aristocracy is parasitic on
the society that it ostensibly serves and that has the power of election over
it. Although some representatives may be very well grounded in their
constituencies, many representatives are far more likely to view their
fellow “aristocrats” as their reference group than their respective electorates. The supposedly mighty citizenry with its power of election over
officials does not have the power to refuse to elect any of them; it can only
turn out the occasional overtly bad apple. The electorate usually does not
have the temerity to overcome incumbents’ advantage. Burke believes
that citizens should be deferential to their aristocratic leaders. Few people
would argue for such social deference today, although there is pervasive
deference to the power of elected officials and to their celebrity, which is
a peculiarly ugly aspect of modern democracies, perhaps uglier and more
pervasive in the United States than in other advanced democracies.
The U.S. Constitution does not explicitly say either that elected officials
have a special status or that they have a duty to represent their constituents. It is essentially a social convention that they should represent their
constituents. Social conventions generally are not morally binding except
through considerations other than the fact that they are conventions.
Moreover, there is no constitutionally determined principle of representation to which elected officials are bound. The range of possible principles is so large and various that all officials could claim to be living up to
one or another principle. For example, a legislator could claim to represent some part of a particular geographical constituency and to support
whatever the leaders of the relevant interest groups in that constituency
want.
Add to various visions of the nature of representation the problem of
citizens’ ignorance. This is a problem that does not bother a Burkean,
because Burke’s arguments are grounded in the presumption of citizens’
ignorance and their deep incapacity to decide what policies their government should adopt. It is also not a problem for Schmitt’s views, because
he supposes that the role of the representative body is not to represent but
to find truth; that citizens do not know truth is taken for granted in the
very structure and purpose of legislative government. In both Burke’s
and Schmitt’s views, elected representatives might suppose that they
should do for citizens what citizens do not even know they want done.
Indeed, citizens might not even want it done once they had given it some
29
John C. Calhoun, A Disquisition on Government and Selections from the Discourse (1851),
ed. C. Gordon Post (Indianapolis, IN: Bobbs-Merrill, 1953), 14–15.
84
RUSSELL HARDIN
thought because, virtually by definition, they are not in the places of the
Burkean or Schmittian representatives who have superior capacities for
determining what should be done.
Although Madison’s views include some of the elitism of Burke and
others, he is, with Manin, more focused on the way things actually work
than on a recommendation that we create a deliberately elitist system.
Mill stands somewhere between these theorists in that he supposes that
elected parliamentarians should be elite in their qualifications but not, as
Burke supposes, via birthright. Mill is also less interested in representation per se, so that he would allow extra votes (which he calls plural
votes) to such people as fellows at Oxford and Cambridge Universities,
because they would likely be better qualified to vote intelligently.30 He,
too, seems more driven by some (extremely vague and underarticulated)
sense of the public good than by concern with representation of varied
interests. Perhaps it is the fact that Madison is engaged in actually creating a representative government that makes him the most realistic and
focused of all these theorists. His realism is driven in part by his concern
to deal with the divisiveness of factions that seemed to fracture the political affairs of several of the states in the United States under the Articles
of Confederation.
Madison wants representatives to represent moderately large communities, so that they will not be too focused on narrow issues. (In this, he
opposes the Anti-Federalists, who want representation down to the small
community level.) But he does not have a conception of “the” public
good, and he does not expect the legislature to work for any such good.
It would, rather, somehow aggregate diverse interests. One might argue
that serving constituency interests would be a way of discovering the
collective good in a Madisonian system. This good would not be the true
good, as Schmitt would want, but only the product of a compromise of
interests negotiated by legislators.
Madison is concerned with at least two overriding issues that he might
have thought of as part of the public good. One of these is simply order,
and the other is a relatively uniform national economic system that would
encourage, or at least enable, economic growth and prosperity by facilitating trade, especially by preventing states from placing restrictions on
interstate trade and by standardizing tariffs on foreign trade. There were
parties who would have preferred to keep the economic system diverse
so that they could free-ride on the anticompetitive practices of others. For
example, the antinationalists of Rhode Island benefited from the high
tariffs that Massachusetts imposed on goods from England and the Caribbean. Traders in Rhode Island imported goods from these places without tariffs, and then transshipped the goods to Massachusetts, where they
could undercut the prices of both domestic and imported goods even
while making a substantial profit.
30
Mill, Considerations on Representative Government, chap. 8.
REPRESENTING IGNORANCE
85
Henceforth, I will focus on what we might call ‘Madisonian’, ‘Humean’, or ‘Schumpeterian’ representatives (or, with more criticism for the
unreality of his vision, ‘Millian’ representatives). I will also discuss (in
Section VB) the nature of representation in the current era of audience
democracy. Although they have intellectual appeal, ‘Burkean’ and ‘Schmittian’ representatives are not part of our political world and we need
not dwell on their conceptions. Madisonian representatives are strongly
driven by their interests, and they work for the interests of their constituents not because they necessarily share those interests, but because
they will be rewarded for doing so. Representatives elected by parties
have an analogous interest in supporting their parties’ positions on
behalf of the parties’ electorates. There appears to be no analogous
sense in which politicians of audience democracy have an interest that
mirrors that of their constituents.
IV. Austrian Social Theory
So far the focus has been on the ignorance of citizens. Another similarly
pervasive and important problem is that central government and its agents
cannot know enough to devise good policies in many realms. Much of the
relevant knowledge is decentralized to smaller organizations and to citizens. The knowledge relevant to governing is, therefore, extremely dispersed. There are things I know that you do not know, and so forth, and
things that each of us knows that no one in government can know. This
fact of the nature of our knowledge is clearly of fundamental importance
for the prospect of a centralized economy. Indeed, much of the main
debate over socialist economic organization during the first half of the
twentieth century was about how demand functions could be determined
so that they could be matched with supply that was set entirely by central
authorities.
Suppose we do not have a centralized economy in anything like the
form that existed in the socialist economies of the former Soviet bloc. We
have central fiscal policy, centrally determined regulations for many activities, and central oversight of civil liberties. Many of these central
determinations of policy seem not only to work reasonably well, but
better than the lack of such central controls. Clearly, central authorities
ought to oversee or regulate many activities, but not others.
The nearest equivalent to Madisonian theory in the twentieth and early
twenty-first centuries has been Austrian economics, as represented by
F. A. Hayek among others.31 Although it is ostensibly an economic theory,
its most cogent insights apply to broader social theory. An especially odd
aspect of the current hegemony of Austrian and Madisonian views of
politics and society —even without these labels attached —is that they
were almost purely theoretical when enunciated. In the past, there was no
31
Russell Hardin, “Seeing Like Hayek,” The Good Society 10, no. 2 (2001): 36–39.
86
RUSSELL HARDIN
way to test Austrian views on the ground. But now they have been and
are being tested, and they seem to be acquitting themselves very well,
although shenanigans at Enron, WorldCom, and so on are attributed by
some to the loosening of government regulation. Madison himself was
not willing to practice his theory once he was in office, and perhaps if
they had gained office, Hayek and others of the Austrian School would
not have been either. But the example of the Soviet world, admittedly a
bad version of socialist statism, compared to the freer but partially trammeled markets of the more prosperous West, gave us a chance to see a
crude, perhaps second-best test of the Austrian-Madisonian theses.
Of course, my assessment is made while we are in the midst of extensive changes, and it might turn out to be grossly optimistic, a mere extrapolation from the most positive aspects of current appearances. But for
the moment, the Austrian and Madisonian schools seem to have the right
vision. This is a stunning turn, even more stunning to those on the traditional Left than to Millian libertarians. The most impressive consequence is the reversal of the long historical trend toward ever increasing
state hegemony over the economy, and all else when the state fell into bad
hands, such as Stalin’s or Hitler’s.
The centerpiece of the Austrian-Madisonian vision is that the knowledge to run a society is widely distributed and most of it cannot become
available to a central government. Hence, a central government should
not attempt to manage society in detail, and it should not attempt a
massive redesign of society. Austrian economists typically worry about
central control of the economy, but they could just as well worry about
central control of social relations more generally. Political scientist James
Scott especially deplores what Hayek calls “Cartesianism” and what Scott
calls the “high modernism” of arrogant redesign of major parts of society,
as in the effort to design cities in supposedly more rational ways,32 or to
reorganize peasants into collective farms in the Soviet Union or into
Ujamaa villages in Tanzania.33
Austrian constraints mean that legislators cannot know in detail what
their constituents want. On average, a member of the U.S. Congress has
a district of nearly six hundred and fifty thousand people, and a senator
from all but a handful of states represents millions of people. These
representatives cannot know their constituents. Survey research at its best
(as it seldom is) cannot determine very clearly what people want from
government or what would benefit them if government acted in relevant
ways. Commonly, when government tries to benefit citizens through attention to their narrow interests, it is not by directly providing benefits,
but only by regulating, prohibiting, or enabling various activities. Of
32
James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition
Have Failed (New Haven, CT: Yale University Press, 1998), chap. 4.
33
Ibid., chaps. 6 and 7.
REPRESENTING IGNORANCE
87
course, there are exceptions, as evidenced by the astonishing scale of
government largesse toward agriculture both in the United States and the
European Union. It is only on relatively big issues that public preferences
are likely to be known, and very often even for such issues —for example,
health care —understanding is radically defective, so that it is virtually
impossible to know what would serve constituents well, much less what
would serve them best.
V. Theories of Representation
The two most rigorous accounts of representative government that are
relevant to actual practice in modern democracies are those of Mill and
Manin. Each focuses much of his discussion on representatives, and each
has some sense of the “Austrian” constraints that I discussed above. Let
us briefly canvass Mill’s and Manin’s theories in order to set up an account of what the functions of representatives should be and, from that,
infer their role morality. I will not provide a full rendition of the views of
Mill and Manin, but only of those aspects that are especially relevant to
understanding the role morality of representatives.
A. Mill
In his Considerations on Representative Government (1861), Mill supposes
that a good government must provide order and progress. For this, a sine
qua non is obedience (but not excessive obedience). The requisites of
order and progress are much the same, because of the dynamism of the
problems that human beings face (387).34 For government to work well,
the most important consideration is the quality of citizens (389). Hence,
the most important tasks of government are to promote the virtue and
intelligence of citizens (390) and to organize what virtue and intelligence
already exist (392). Of course, Mill is a welfarist, and therefore his central
claim is that government must enable the people to do well. For this and
for progress more generally, liberty and individuality are fundamentally
important (396–97).35
Mill notes that it is historically commonplace to assert that the best
form of government is benevolent dictatorship (chapter 3). He says on the
contrary that, for example, the benevolent despotism of Augustus set up
Romans for the “more odious” reign of Tiberius (403). Moreover, an autocratic government cannot know enough (399) to run the society well.
Mill essentially presumes an Austrian social theory. He also anticipates
34
In this section only, citations to pages and chapters in Mill’s Considerations on Representative Government will be given in parentheses. See note 11 for complete bibliographic
information.
35
Mill gives a compelling consequentialist justification for liberty, without which progress
is eventually stifled.
88
RUSSELL HARDIN
Schumpeter’s main point, that a voter’s incentive is to be ignorant because, as Mill puts it, “a person must have a very unusual taste for
intellectual exercise in and for itself who will put himself to the trouble of
thought when it is to have no outward effect” (400). Mill holds an essentially pragmatic view of knowledge. He also reinforces this view when he
says that no one will take an interest in government who cannot participate (469). This suggests that almost no one will take much interest in
government, because almost no one can participate to any significant
extent. But Mill’s point in this passage seems to have been intended to
suggest that people could be motivated to participate through education.
Unfortunately, education might sooner lead one to understand just how
massive are the obstacles to real participation most of the time. Mill
himself concludes that there is very little opportunity for holding office in
the central government (chapter 15). Local government, with its smaller
scale and higher proportion of office holding, helps (535–36), but even
local government in modern democracies is commonly carried out by
representatives who have constituencies numbering in the thousands and
even hundreds of thousands, so there cannot be very much participation
even here.
In a continuation of his quasi-Austrian views, Mill says that what a
legislative body is competent to do is quite limited (chapter 5). It can
adopt not very precise legislation —for example, in our time, tax policy or
environmental regulation. To do much more, it must delegate such things
to expert advisers and to committees (430). Even then, laws will tend to
become inconsistent over time because no one will have massive oversight of the whole body of law (a task performed by the Nomothetae in
ancient Athens [431]). Hence, the proper function of a legislature is the
very limited one of watching and controlling the administrative branches
of government (432).
Mill famously claims that it is especially useful to engage in debate
(433). This is not the “deliberative democracy” of current visions, although Mill’s authority is often invoked in its support. His claim may
partly suggest Schmitt’s concern with deliberation as a device for finding
the truth. Against Schmitt, however, Mill thinks that genuine representativeness is required, because it enables the legislature “to indicate wants,
to be an organ for popular demands, and a place of adverse discussion for
all opinions relating to public matters . . . ; and, along with this, to check
by criticism, and eventually by withdrawing their support, those high
public officers who really conduct the public business” (433).
Mill attributes the infirmities of representative government to two general causes (chapter 6). First, there is the quasi-Austrian general ignorance
and incapacity of government and its agents; second, there is the danger
of being under the influence of interests not identical to the general welfare of the community (436). He supposes that the main comparison to be
made is between government by bureaucracy and government by repre-
REPRESENTING IGNORANCE
89
sentative democracy (438–39). Autocracy and aristocracy are too obviously flawed to merit consideration.
Against the danger of class legislation, Mill optimistically supposes
that the few individuals who are public-spirited combined with those
whose own interest happens to coincide with the public interest can
collude in a majority to carry the day for the public interest (446–47). This
is Panglossian. He should have read Madison more attentively. Indeed, he
should have read himself more attentively.
In chapter 8 Mill discusses the extent of the suffrage and his view that
it should generally be extended and should include women (479–81). Mill
argues that fully representative democracy is the only true democracy
(467). In keeping with his general view that citizens must be required to
be intelligent enough to see their own interests if they are to gain the
franchise, he proposes educational requirements on suffrage, because voters must be able to read, write, and do arithmetic if they are to vote
intelligently on their own interests. For him this seems to be little more
than a definitional implication of the idea of democracy: universal teaching must precede a universal franchise (470).
Mill especially worries about the fact that the large majority of citizens
in his England are manual laborers. He supposes that they might abuse
their power through their democratic majority to enact class legislation.
He was largely wrong about this empirical point, because manual workers in his time were a diverse lot with conflicting interests. Manual laborers included farmers as the largest group, factory workers (who only
became the bare majority of the English work force several decades after
Mill wrote), and a miscellany of service workers, especially in urban
areas. (In the United States in his time farmers were the overwhelming
majority of all workers.) These quite diverse groups had little in common.
Indeed, the only interest that they might have shared was in redistribution from the middle and upper classes, which in any case was not a
viable political program in Mill’s day. (Madison, too, worries about the
possibility that the poor majority could combine to dispossess the wealthy
of their property, a worry that made some sense empirically in his time,
but not in ours, when evidence says it does not happen.)
Mill advocates universal but unequal suffrage (473–74). Uniquely
among democratic theorists up to his time, he proposes intelligence,
not property, as the qualification for extra weight (478). He grants,
however, that unequal suffrage may not be a practical suggestion (476).
He is also inclined to contrive devices that give greater weight to some
voters over others, as through the use of gerrymanders to block workers from achieving a parliamentary majority by concentrating their votes
in certain districts that they would win overwhelmingly while diluting
their strength elsewhere (477). (This is what he calls a “fully representative democracy.”) He gives a not very convincing causal claim in
support of such voter inequality. He says that the best incentive to the
90
RUSSELL HARDIN
growth of intelligence is rising to power, not having achieved it (479).
This claim is a pointless fallacy of composition: the class might rise to
power but almost none of its members would. Thus, although the class
of workers might rise to power, individual workers generally cannot.
Mill thinks that a constitution should not declare ignorance to be entitled to as much political power as knowledge —as the U.S. Constitution
virtually does declare (478). He therefore excoriates the American system.
B. Manin
In the first two chapters of his The Principles of Representative Government, Manin gives an account of the great transformation from direct to
representative democracy.36 In Athens, there was general representation
of all by all in certain bodies of government, coupled with selection by lot
of people to serve in more restrictive bodies. Now democracies are with
rare exception systems of elected representative bodies. For Madison and
Emmanuel-Joseph Sieyes, the great constitutional thinkers of the American and French democratic revolutions, respectively, representative democracy was a new and, for its time, preferable form of government.
Manin’s main thesis on the development of representative government
is that it inherently tends toward aristocracy of an odd and familiar kind.
Membership in the class of aristocrats is determined by citizens through
elections, and the characteristics that earn entry into this class have changed
over time, from something like social distinction in Madison’s time and
earlier in colonial America and England, to something more nearly like
celebrity today. This aristocratic tendency of elected representative government was already foreseen by the ancient Greeks, who preferred to
select officeholders by lot, considering it more genuinely representative
(27). By lot, even the lowliest citizen could be selected for a government
position. Montesquieu famously observed that there is a close link between lot and democracy and between election and aristocracy (70). Manin says that rotation in office and selection by lot reflect deep distrust of
the professionalism that would follow from specializing in holding office
over a long period. Madison and Sieyes, however, want professionalism
(32). They worry that democracy could entail putting power in the hands
of amateurs (33). They believe that officeholders should be more distinguished than the ordinary run of citizens (94). In actual democracies,
wealth, property, and the payment of taxes have often been employed as
qualifications for running for office (97–98), and wealth in our time is
often still very useful for gaining office by spending lavishly on one’s
campaign.
36
In this section only, citations to pages and chapters in Manin’s The Principles of Representative Government will be given in parentheses. See note 6 for complete bibliographic
information.
REPRESENTING IGNORANCE
91
In the Constitutional Convention of 1787, and in the state ratifying
conventions and the debates leading up to them, the Anti-Federalists held
that representatives should fairly closely mirror their constituents (109).
Representation was not a matter of giving representatives mandates but
of having the diverse U.S. population represented by their own types
(109). Anti-Federalist Samuel Chase pointed out that the great majority of
the population —farmers and ordinary workers —could never be elected
(112). Before early fascist or corporatist guilds, the Anti-Federalists were
probably the only true advocates of genuinely representative democracy
by station in life rather than by geographical location, and there has been
no further debate on the aristocratic nature of election after the AntiFederalist arguments of 1787–88 (132). Manin concludes that, for solid
causal reasons, “election cannot, by its very nature, result in the selection
of representatives who resemble their constituents” (149). In the end,
however, power is not earned by distinctive traits or capacities but by
agreement among the electorate about what traits constitute superiority
(158). Apart from his being the son of a former president, one would be
very hard pressed to account for the traits that put George W. Bush in the
presidency, but one can give an account of how support coalesced around
him at various stages in the 2000 election cycle. Still, Manin notes that the
principles of distinction and salience that make for election do not violate
norms of equality and political right, although the constraint of wealth
does (159). Mill thought that American political leaders in his time were
woefully undistinguished.37
In the traditional view, members of the British Parliament represented
the nation as a whole and not merely their constituencies (163), as in the
views of Burke and, to some extent, Mill. This view has given way, first
to party democracy, in which a party gains control of parliament and
governs for its term (206–18), and then to what Manin calls “audience
democracy” (218–32), which is substantially formed by the media and the
capacity of individual candidates to appeal to the voters.
Against Schmitt’s view that deliberation is valued for giving us access
to truth, Manin concludes, “It is the collective and diverse character of the
representative organ, and not any prior or independently established
belief in the virtues of debate, that explains the role conferred on discussion” (187). What makes some resolution of legislative debate a law is
some form of consent, not discussion of it or its truth value (189). Moreover, the requisite consent is typically merely majority agreement, whereas
truth should command unanimous consent once it is established.
In keeping with Manin’s thesis on audience democracy and “celebrity,”
the great playwright Arthur Miller (1915–) analyzes the acting abilities of
37
John Stuart Mill, “De Tocqueville on Democracy in America, II,” in John Stuart Mill,
Essays on Politics and Society, vol. 18 of Collected Works of John Stuart Mill, ed. J. M. Robson
(Toronto: University of Toronto Press, 1977), 175.
92
RUSSELL HARDIN
modern American presidents, going back to Franklin Roosevelt.38 Miller
evaluates the acting ability of these presidents as he would assess how
and why an actor succeeds on the screen or on the stage. He especially
notes the peculiar differences between live and screen performances; in
the latter, close-up cameras could turn more intense expressions into
something baroque or rococo. In the television age, candidates must be
flat; they cannot be orators in the grand style of, say, William Jennings
Bryan. Miller argues that successful politicians tend to master performing
before the camera and that we the voters value them in part for this
success. Miller’s analysis is far more sophisticated in its appreciation of
the theatricality of politics than we are accustomed to from pundits, who
typically lack his professionalism. Miller himself came to appreciate the
difference between screen and stage from participating in the filming of
The Misfits (1961). Clark Gable, the film’s star, explained to Miller that he
had to play his part very low key. Roosevelt might partly be Miller’s
favorite from the cast of modern presidents because he politicked in the
era before television, which allowed him a florid acting style, rather than
the flat affect favored by presidents in the television era.
VI. The Role Morality of Representatives
Carl Schmitt holds that democracy is the identity of ruler and ruled,
and this is not compatible with representation; Rousseau believes that
representative democracy is slavery.39 Schmitt’s is merely a definitional
move. If we suppose Schmitt’s identity, it is pointless for us to talk about
the role morality of the rulers with respect to the ruled. The issue only
arises because there is not an identity between the ruler and the ruled in
a representative democracy. Similarly, if representative democracy is a
form of slavery, it is silly to speak of a role morality for the overseers of
the slaves. Against both these views, elected officials are, by a complex
formula, both the agents and the rulers of the citizenry.
My purpose is to analyze the role morality of elected representatives in
the light of the more credible theories of representation that I have canvassed: Madisonian constituency representation, party representation, and
representation in Manin’s audience democracy. I propose that we define
the morality of representatives as artificial duties derived from their roles.
There are two other ways that we might proceed. We might apply
something like conventional morality to the roles of representatives.
Apart from intuitionists, no moral philosopher would do this, but we
might suppose that, empirically, there is a broadly expected —therefore
conventional —morality for elected officials. One measure of such a con38
Arthur Miller, On Politics and the Art of Acting (New York: Viking, 2001).
Manin, The Principles of Representative Government, 151, 1. Rousseau specifically discusses England in the eighteenth century. He wrote, of course, before the rise of representative democracy in its fuller forms.
39
REPRESENTING IGNORANCE
93
ventional morality might be gleaned from surveys on why people vote
against someone whom they previously supported. I doubt that there
is a standard, widely accepted conventional morality for elected officials, but there might be, or there might have been in some eras. Insofar as there is such a conventional morality, it seems likely to be related
to the functional role moralities that I will canvass below.
Although it seems unlikely that there is even the hint of a consensus on
any broad conventional morality for elected officials in modern democracies, there might be nearly a consensus that our officials should not use
their offices to work for their private benefit against the interests of their
constituencies or of the larger public. There is a standard moral constraint
on agency relationships in general, which bars the agent from taking any
action that would be a conflict of interest. Agents should not use their
position of acting on the authority of others to take advantage of them.
While this principle is seemingly simple in the contexts of many agency
relationships —in our dealings with lawyers, doctors, accountants, and so
forth —articulation of such a principle for elected officials is complex, as
I shall argue below.
A. Functionally determined morality
The role morality of a doctor is to see to the patient’s health because this
is the function for which the doctor’s services are sought; the role morality of a lawyer is, analogously, to see to winning the client’s case or
giving beneficial legal advice. If we wish to determine what the role
morality of elected officials is —by analogy to professional ethics —we
must first settle on what the function of an elected official is in relation to
a constituent. This depends on our explanatory (not normative) theory of
representation. Clearly, there cannot be a generally correct role morality
for political representatives in the way that there is for doctors, because
there are many theories of representation and these require different principles of action by elected officials.
We have at least three practical theories of representation that are quite
distinct from one another. First, there is the quasi-Madisonian theory that
focuses on individual legislators as agents of their constituencies. Second,
there is the political party theory that makes elected officials the agents of
their parties, and parties the agents of their partisan constituencies, which
typically are broadly defined classes rather than geographical constituencies. Third, there is Manin’s audience democracy, in which it is not clear
that elected officials are agents of anyone other than themselves. The first
two theories yield relatively straightforward principles of role morality
for representatives. Historically, the role morality that seems most commonly stipulated by citizens in many democracies is that representatives
be seen as agents of their constituents or as party loyalists. Given that
representatives cannot know in detail what their constituents’ wants are,
representatives can knowledgeably only address their constituents’ inter-
94
RUSSELL HARDIN
ests as fairly broadly conceived. Therefore, the role morality of an elected
lawmaker is to see to the broadly defined interests of his or her constituents. We can call this the ‘Madisonian role morality’ of elected legislators, which had its greatest influence in the early United States. We can
call it ‘party role morality’ in systems organized by parties, although in a
party system, service to one’s constituency is rendered indirectly through
service to one’s party, which serves all relevant constituencies.
If we consider the current stage of development of representation, Manin’s audience democracy, we may wonder whether any role morality
still applies to elected officials. One might assert a role morality, but the
voters are unlikely to be concerned with it, except perhaps when a representative is grossly out of line. Suppose we do not elect our officials on
the ground of how we expect them to handle their role in office, but
primarily on the ground of who they are, including whether they happen
to be celebrities. Then they would seem to have no mandate beyond
continuing to be themselves. If media mastery is the route to election,
then it might also be the route to renown while in office. It would be
perverse for the electorate to complain that their officials are very good at
precisely the skills that got them elected.
We might still think that the role morality of representatives is to represent, to work as agents for their constituents’ interests, sometimes to
seek the larger public good, and sometimes to defend civil liberties. But
in America our moral expectations evidently do not explain or even correlate with how we vote to select our representatives. It was a happy fact
of both Madisonian individual and later party systems that our interests
coincided with our moral expectations for our representatives. In audience democracy, they do not correlate well.
Voters sometimes do hold candidates accountable for lack of media
appeal, as in the case of Al Gore and George Bush in 2000, or for loss of
it, as in the case of Gary Condit in 2002. Gore was often ridiculed for his
stiff, dull manner (with at least one popular comedian comparing him to
a wooden cigar store Indian). Bush, who barely won a close election over
Gore, had likewise been satirized for his frequent verbal clumsiness. Prior
to 2002, Condit had been reelected several times, even though his legislative impact was nil throughout his time in office. When he became the
furtive man who refused to speak to the press about his reputed affair
with his murdered former intern, his career was doomed, even if it could
have been shown that he had played no role in her murder, and even
though mere revelation of his affair, had the woman not been murdered,
might not have blocked his continued reelection. By appearing indifferent, if not sinister, he simply lost much of the personal appeal that had
repeatedly won him reelection.
If personality and media success are the grounds for election, then the
electorate is likely to be divided on most politicians just because different
personalities are likely to appeal to different voters. It seems probable, for
example, that the astonishing divergence of views on Bill Clinton had
REPRESENTING IGNORANCE
95
much to do with divergent responses to his personal style. Even as he
shifted the Democratic Party toward the center of American politics, he
was detested by the American Right as perhaps no one since Franklin
Roosevelt. In keeping with Manin’s thesis on the evolution from Madisonian to audience democracy, Roosevelt seems to have been despised
for his political positions, Clinton for being Clinton.
The development of audience democracy raises the question of whether
personality might correlate with political positions to a sufficient degree
as to yield the cues that Popkin and others need for their argument that
voters base their votes on low-information rationality. There are two interlocking sets of correlations: first, on the side of the candidates and,
then, on the side of the voters. Candidate personality must correlate with
candidate position, and, given this correlation, candidate appeal must
correlate with the voters’ positions on issues. These correlations are tenuous reeds for us to hold onto in the hope for rational politics, if we have
indeed entered an era of audience democracy. Against even such a slim
hope for rationality, it seems likely that audience democracy is a response
to the combination of voter ignorance of policy and government and of
politicians’ ignorance of voters’ interests other than in broad terms. If we
put the Schumpeterian vision of the limited capacities of voters together
with the Austrian vision of the limited capacities of the state, we have
representatives who cannot know much about their constituents trying to
represent constituents who do not even know their interests in many
areas and who, in any case, know very little about their government, its
policies, and its officials.
B. Conventionally determined morality
Beyond any role morality that is functionally determined by our theory
of representation, we might insist on several other “moral constraints” (that
derive from conventional morality) on what our representatives do in office. There is one fairly broad, general concern that representatives are commonly expected to address: the political equality of citizens. This is a concern
that seems to follow from the nature and purpose of democracy. We might
fundamentally disagree about the extent to which economic inequality is
good (as in the theory of justice of John Rawls, who allows inequalities —
which might be extreme —that redound to the benefit of the worst off ), but
we do not generally argue in public that political inequality is good, although some, including Mill and Texas oilman H. L. Hunt, are notable exceptions. Exactly what it takes to make individuals politically equal is not
easily determined, but some elements seem clear enough.40 Anything that
is an obstacle to political participation, such as extremely poor education,
and that might be affected by public policy is an issue that we might ex40
For discussion, see Thomas Christiano, The Rule of the Many: Fundamental Issues in
Democratic Theory (Boulder, CO: Westview Press, 1996).
96
RUSSELL HARDIN
pect our representatives to take on, even though it goes beyond our own
interests and beyond their representation of our interests.
As noted earlier, Mill supposes that two of the most important tasks of
government are to promote the virtue and intelligence of citizens, and to
organize what virtue and intelligence already exist.41 These two functions
are of such salience for him because he supposes that the most important
consideration in creating and running government is the “quality” of
citizens,42 which might be partly a concern with political equality. (Against
such generosity of interpretation, however, we know that Mill strongly
defends inequality of political power grounded in intellectual qualifications.) If we thought that citizens played a substantial and active role in
government, then we might agree with this claim, but it is prima facie
false that citizens play a great role. They might occasionally mobilize
effectively and bring about a change in government policy, as may have
happened in the civil rights and anti–Vietnam War movements in the
United States. But such activism is surely a rare activity for citizens, who
generally attend to their own lives, acquiesce in government’s discretion
to determine and carry out public policy, and even acquiesce to the extreme as millions are sent off to fight and die in “great” wars.
One might contrive an argument that, say, broader education would
redound not only to the benefit of those educated by this new policy, but
also to the benefit of more or less all citizens through the creation of a
more productive or otherwise more appealing society. But in general, it
seems likely that a representative who works for such an egalitarian
policy for the whole society will risk harming the interests of his or her
constituents, who may not directly benefit from, but may directly pay for,
the programs that enhance educational equality. Some things that a representative might seek for his constituents can most readily be attained if
they are provided for all those in the relevant class. For example, the best
way to guarantee a minimum wage for my constituents might be to
legislate it for the entire nation. Hence, I might act in a way that is similar
to the actions of a Burkean or Millian representative concerned with the
general welfare and not merely with the welfare of my own constituents.
We could stipulate other “moral constraints” on representatives, such
as seeing to the constituents’ moral development or their religious beliefs.
These two purposes were ruled out by the American constitutionalists,
although they have been stipulated to be a large part of the mandate of
government by many other regimes. In American politics there have been
major movements that advocate the use of government for other purposes, for example, the social-agenda crowd that wants regulation of
values, the mercantilist-statist Right that wants government to protect
business interests, or the socialist-statist Left that wants government to
restructure society to achieve greater economic equality.
41
42
Mill, Considerations on Representative Government, 390, 392.
Ibid., 389.
REPRESENTING IGNORANCE
97
Additionally, we can probably claim that part of the role morality of some
government officials in the United States is to defend individual liberties,
such as those defined in the Bill of Rights. For many officials, such as legislators, the defense of liberties requires merely refraining from infringing
them, although, on occasion, legislators might be called upon to devise new
protections, and other officials, especially those in the justice system, might
be expected to defend liberties against official abuse. It is primarily the
judiciary that has the function and, therefore, the role morality to defend
civil liberties. Elected officials and appointed officials, such as the U.S.
Attorney General, frequently find civil liberties an annoyance.
Finally, we might wish to press upon legislators as part of their role
morality the more general purpose of working for the interest of the
entire public, and not merely for their constituents. In a variant on my
example of education, many legislators who represent districts in which
education meets reasonably high standards might be expected, nevertheless, to vote for legislation mandating such standards for the entire nation, not in order to enhance political competence (as in my original
example), but to enrich the life prospects of those who would benefit from
the new programs. The votes of these legislators would serve not the
interests of their own constituents, but only, in some sense, the broader
public good or even only the interests of other citizens outside their
constituency, possibly even at a substantial cost to their own constituents.
If the policy were seen as a public good, then support for it could easily
be justified on Burkean or Millian grounds, but it might sometimes also
be argued on the ground of concern for political equality, in which case it
would be an outgrowth of a conventional morality.
Clearly, none of these conventional moral principles for legislators is
strongly backed by the electorates of many democratic nations. These are
in many cases idiosyncratic views endorsed by activists of various stripes,
not views inferred from the logic or nature of representation. Some of
them might be backed by particular moral theories, such as utilitarianism,
which might also back the institutions of representative government, but
the principles are still not inherent in the nature of representation.
C. Conflicts of interest
Finally, we ought to consider one aspect of the role morality of anyone
who acts as an agent for anyone else. All agents must avoid conflicts of
interest that could lead them to benefit themselves at the expense of their
principals. If our elected officials are our agents, then they must adhere to
this constraint. The problem of conflicts of interest raises what is apparently the most striking difference between traditional professional ethics
and any plausible ethics for elected officials: both the traditional professional and the elected official are agents on our behalf, and we want them
to act in our interests, but the politician’s position is far more complex
than that of the traditional professional. For example, every doctor to
98
RUSSELL HARDIN
whom I go as a patient should be my agent. But a representative can be
seen as my agent and not yours in a meaningful and important sense if I
voted for this representative while you voted against her. You cannot
claim that she should work for your interests in the same way that I can,
because she should act for the majority who elected her, including me,
more than for those who opposed that majority.
Moreover, you might want to have her removed from office if she engages in practices that seem to benefit herself at public expense. I might
not want her to be removed because I suspect that any replacement would
be less committed to serving my interests. Consider the complications that
might factor into opposition or support for a politician aside from the merits of any charges against him. As a real-world example, consider the positions of those who supported or opposed the impeachment of President
Clinton and his removal from office. (The concern in his case was not over
any abuse of office for his own interest.) Obviously, there was a substantial
correlation between one’s position on these moves and one’s interest in having Clinton continue in office. Similarly, supporters of Berlusconi and his
Northern League have argued that court actions to try him for bribery are
politically motivated and opportunistic and, as I previously noted, they support legislation that would protect him from prosecution.43
If we are legislating in advance, when there is no political valence to
corruption, then we might all agree that acting from certain classes of
conflict of interest should be punished by removal from office. When
there is an actual case, you might favor removal while I do not, because
your interests are served by removal and mine are not. The role morality
for the officeholder will then be whatever we have stipulated by law and
will not be colored by our own interests in removal or retention of a
particular person in office. We could sensibly say that there is a public
interest in blocking certain classes of conflict of interest.
VII. Conclusion
If democracy were strictly representative, then government would be
an epiphenomenon determined by the wants and interests of citizens, as
in the view of political scientist Arthur Bentley (1870–1957).44 It is not
merely an epiphenomenon, though, because citizens’ interests and demands are only weakly determinative. Government takes on a life of its
own that has much to do with the elevation of political leaders to a
peculiar aristocracy. This aristocracy is not the oligarchy of Michels, because its members are far more subject to election, and they can occasionally be unelected by the larger public. Michels’s oligarchs were subject
to control only within their organizations, not by the larger public.
43
Bruni, “Italy, a Land of Tolerance.”
Arthur F. Bentley, The Process of Government (Chicago, IL: University of Chicago Press,
1908).
44
REPRESENTING IGNORANCE
99
A saving grace of aristocratic representative democracy is that, in any
case, democracy works at all only where there is fairly broad consensus
on political order and, commonly, civil liberties. This consensus means
that most citizens need merely acquiesce in allowing government to run
or intrude into large parts of their lives. As political scientist Robert Dahl
says, “In a sense, what we ordinarily describe as democratic ‘politics’ is
merely the chaff. It is the surface manifestation, representing superficial
conflicts. [These] disputes over policy alternatives are nearly always disputes over a set of alternatives that have already been winnowed down
to those within the broad area of basic agreement.” 45 This is roughly
Tocqueville’s view as well: “When a community actually has a mixed
government —that is to say, when it is equally divided between adverse
principles —it must either experience a revolution or fall into anarchy.” 46
We should qualify Dahl’s claim with the note that “the broad area of
basic agreement” need only be an area in which the politically effective
groups are in agreement. Indeed, it need merely be an area in which the
aristocratic political class is in agreement while the rest of the population
basically acquiesces.47 We might revise the Dahl and Tocqueville view to
fit current conditions and say that much of the chaff of politics today is
more nearly a part of the image than of real policy-oriented concern in its
own right. For example, the chaff of Gary Condit’s life dominated the
media for weeks until it was reduced to its properly trivial status by the
terrorist attacks of September 11, 2001.
Finally, note the irony that it is only because citizens began to be somewhat educated that representative democracy could arise in a large state
(although the state’s capacity to take a census and its technological capacity to collect votes have also played roles).48 Yet it is the limits of
citizens’ understanding that makes it difficult for us to assess the quality
of the very representatives we elect and, in particular, to determine whether
they live up to any role morality we might assign to them. In the face of
current trends in electoral motivations, it seems unlikely that the electorate consistently has in mind any role morality, either conventional or
functional, for the media masters we put in office.
Politics, New York University, and Political Science, Stanford University
45
Robert A. Dahl, A Preface to Democratic Theory (Chicago, IL: University of Chicago Press,
1956), 132–33.
46
Alexis de Tocqueville, Democracy in America, 2 vols. (1835, 1840; reprint, New York:
Knopf, 1945), 1:260.
47
Russell Hardin, Liberalism, Constitutionalism, and Democracy (Oxford: Oxford University
Press, 1999), chap. 4.
48
Scott argues that a state’s capacity to keep records on us or, in his term, to make us
‘legible’ allows the state to control us in various ways, such as by raising taxes and armies.
This capacity also makes it possible to determine just who is to be represented in a modern
democracy. Indeed, the U.S. Constitution requires a periodic census primarily for the purpose of allocating seats in the House of Representatives according to state populations so
that representation is relatively equal.